[linux-dvb] cx18: Testers needed for patch to solve non-working CX23418 cards under linux (Re: cx18: Possible causal realtionship for HVR-1600 I2C errors)
loudestnoise at gmail.com
Mon Oct 6 04:17:31 CEST 2008
Well since lspci -v still doesn't recognize my HVR-1600 after
switching around card positions and even removing my other tuner card
should I assume I have a bum card?
>> >> I can't seem
>> >> to get the card to initialize. dmesg | grep cx18 returns nothing.
>> > OK. Let me ask some basic questions that you may have already verified:
>> > 1. Does the following command show a cx18.ko module installed?
>> > $ find /lib/modules -name cx18.ko -print
>> Yep, sure does.
>> > 2. Does this command:
>> > $ /sbin/lspci -v
>> > show a
>> > "Multimedia video controller: Conexant CX23418 Single-Chip MPEG-2 Encoder with Integrated Analog Video/Broadcast Audio Decoder
>> > Subsystem: Hauppauge computer works Inc. Unknown device 7444"
>> > ?
>> Nope, this concerns me. I do see stuff about my PVR-500, but nothing
>> that says Conexant CX23418.
> Well, if lspci can't see the card, no amount of mucking with the cx18
> driver is going to help.
> See if
> $ /sbin/lspci -n
> returns something with "14f1:5b7a" which is the PCI vendor and device id
> for a Conexant CX23418, like this:
> 03:03.0 0400: 14f1:5b7a
> or a number that is very close (maybe wrong in one or two bit
> If it doesn't show at all, invoke lspci as root to in mapping mode
> (dangerous, according to the man page) to see if it finds the device:
> # lspci -n -H1 -M
> Otherwise, you'll have to take other measures (pulling cards, moving
> into a different slot, etc.) to see if you can get the board recognized
> by the PCI subsystem.
> If a PCI ID that is close, but wrong in a bit position or two, shows up,
> then pull the card, blow the dust out of the slot, put the card back in
> the machine, and try again.
>> > does modporbe emit any errors and what does 'dmesg' or /var/log/messages
>> > show? (Please don't grep on cx18, as not all relevant messages will
>> > have cx18 in them.)
>> No error messages on the modprobe, and here is the portion of dmesg
>> that talks about my cards and things like ivtv and such.
>> 27.820719] Linux video capture interface: v2.00
>> [ 27.831881] ivtv: Start initialization, version 1.4.0
>> [ 27.832020] ivtv0: Initializing card #0
>> [ 27.832024] ivtv0: Autodetected Hauppauge card (cx23416 based)
>> [ 52.632670] ivtv1: Loaded v4l-cx2341x-enc.fw firmware (376836 bytes)
>> [ 52.831513] ivtv1: Encoder revision: 0x02060039
>> [ 56.375834] cx25840 1-0044: loaded v4l-cx25840.fw firmware (16382 bytes)
>> [ 62.593527] ath0: no IPv6 routers present
>> [ 953.888579] cx18: Start initialization, version 1.0.0
>> [ 953.888979] cx18: End initialization
> The cx18 driver loaded up, the PCI bus was scanned for matching devices,
> and none were found apparently. No errors, but also no card with the
> proper PCI ID visible on your PCI bus.
>> [ 1677.647652] i2c /dev entries driver
>> [ 2149.083630] ivtv1: ================= START STATUS CARD #1 =================
>> [ 2149.083639] ivtv1: Version: 1.4.0 Card: WinTV PVR 500 (unit #2)
>> [74682.713469] ivtv1: ================== END STATUS CARD #1 ==================
>> [164887.647546] cx18: Start initialization, version 1.0.0
>> [164887.650039] cx18: End initialization
> More of the same.
>> > 5. When the module is loaded, it should always blurt out at least
>> > "cx18: Start initialization, version 1.0.0" in dmesg and messages,
>> > unless the kernel couldn't load the module. If the kernel couldn't load
>> > the module, it should log an error message: what is that message?
>> An error message from somewhere besides dmesg? Let me know where this
>> log resides and I can include that too.
> No. There shouldn't be an error from the kernel in this case (but it
> would be in dmesg), as the module loaded properly and blurted out "Start
> initialization, version 1.0.0". The PCI subsystem doesn't appear to be
> recognizing your hardware. That is not an error condition from the
> perspective of the kernel nor the cx18 driver per se, but it certainly
> is a problem.
>> - David C. (loudestnoise)
More information about the linux-dvb